1100D cold finger sensor cooling with TEC and water cooling

[Gina]

I now have all the cooling bits back in the camera and ready for action. Going out shortly to put the camera on the scope, connect up the water pipes and fill the system.

[Gina]

The cooling is working very well I can get EXIF T down to -22C with short exposures and -17C after a long run with live view on - running sensor and image processor continuously. Ambient temperature was 19C. EXIF T without cooling was 27C. So that's a temperature (EXIF T) reduction of 49C with the cooling on

BUT I have a condensation problem. Quite severe condensation on the sensor which turned to ice as the temperature went down.

Two things I need to do :-

1. Control the cooling - have the parts, just need to make up the circuit.
   
2. Sort out the condensation/icing problem. I already have a couple of fresh silica gel bags in the camera but the case sealing needs improving.
   

I'm not sure whether it will be possible to seal the case as it stands, there are so many controls and other places where the case can breathe. Then there's the front - the only thing stopping moisture getting in is the CLS-CCD clip filter, and that's not sealed. Might stop the worst of the dust but not moisture.

Might be able to enclose the camera body in a polythene bag with seals for the wires and tubes. As for the light path, the only thing I can think of is Baader transparent film. This is the film Baader use to form the base of their solar filter material - totally transparent and very thin and optically perfect. There is nowhere to use a screw-in filter - the camera attaches to the FF/FR with it's bayonet. And the bayonet itself will leak air/moisture.

[nightvision]

That's an impressively low temperature, is that with the one external TEC?. If you can sort the condensation you should be able take very long subs (if you don't have a bad LP problem).

I am about to start on my 1000D but not much point in aiming for such a low temp as I have plenty of LP (that's my excuse for not being as bold as you ), probably just do a sealed ally box. Also living near the sea means humidity is often very high.

Tony.

[Gina]

That's an impressively low temperature, is that with the one external TEC?. If you can sort the condensation you should be able take very long subs (if you don't have a bad LP problem).

Thank you No, that's two stacked TECs inside the camera.

I am about to start on my 1000D but not much point in aiming for such a low temp as I have plenty of LP (that's my excuse for not being as bold as you ), probably just do a sealed ally box. Also living near the sea means humidity is often very high.

Tony.

LP is low here in rural Devon though there is some LP low in the sky to the SW from Honiton plus a very small amount low down in other directions from villages and security lights. Street lights get filtered out by the CLS clip filter but it doesn't help with white security lights.

A sealed ally box is probably a good option for you with high humidity. I got quite reasonable results with a die cast ally box, a 12v 50W Peltier TEC with a large heatsink and fan from an old PC processor plus a small fan inside to circulate the air. The whole thing was enclosed in 25mm thick polystyrene foam insulation. Trouble with that setup was the weight but if you make up your own box from thinner ally sheet (the die cast box is thick and heavy) you should be able to reduce the weight. A heat sink with lots of thin fins is best for cooling.

[yesyes]

The cooling is working very well I can get EXIF T down to -22C with short exposures and -17C after a long run with live view on - running sensor and image processor continuously. Ambient temperature was 19C. EXIF T without cooling was 27C. So that's a temperature (EXIF T) reduction of 49C with the cooling on

That is very impressive. I'm still struggling to get below zero on the far side of the cold finger.

[Gina]

That is very impressive. I'm still struggling to get below zero on the far side of the cold finger.

Thank you Must say, I'm very pleased with it.

Is that with just one Peltier TEC? Using two makes a big difference.

[yesyes]

I've tried with 2 last Saturday for the first time. I've got a 30x30mm 60W and on top of that a 40x40mm 90W, both 12V. The coldest I got on the far side of the cold finger was +6°C. I did blame that on the hot weather (30°C ambient).

I'm using a CPU heatsink with 4 heat pipes, not a water cooler. But the heatsink barely goes above ambient with the fan running. So I take it it is sufficient. It does get quite hot quite quickly when I switch off the fan.

I think I'm still doing something wrong with the way I'm testing. Just can't think of what it might be.

Cold finger too long? Too exposed to the open air?

[Gina]

I think the Peltiers you're using are too powerful. I tried two powerful TECs and got very poor cooling. I'm now using a 5v 20W (approx) and 12v 50W. With that combination on the long cold finger I got -19C at the cooler end of the cold finger and an EXIF T of -9C. So I was losing 10C in the cold finger. The open end was uninsulated and soon got covered with a thick layer of ice.

I was considering going for even lower power TECs as there's really not much heat to get rid of from the sensor but since the present system is working well I shan't bother. I was thinking of a 3.5v 10W 20mm square TEC on the cold finger and a 5v 20W 25mm square one on top. The PC PSU has 3.3v, 5v and 12v so I could run the 3.5v TEC off the 3.3v supply.

[Gina]

To answer your post... Yes, it would certainly seem that your air cooling will be sufficient, particularly if you go for lower power devices like me. A long cold finger certainly looses cooling - I found a big difference by using a short one. As shown earlier in this thread, it's folded over onto the top of the imaging board.

[nightvision]

........

A sealed ally box is probably a good option for you with high humidity. I got quite reasonable results with a die cast ally box, a 12v 50W Peltier TEC with a large heatsink and fan from an old PC processor plus a small fan inside to circulate the air. The whole thing was enclosed in 25mm thick polystyrene foam insulation. Trouble with that setup was the weight but if you make up your own box from thinner ally sheet (the die cast box is thick and heavy) you should be able to reduce the weight. A heat sink with lots of thin fins is best for cooling.

That's the way I was thinking of going, using thin ally sheet and foam like Gary Honis http://ghonis2.ho8.com/rebelmod450d16c.html . Maybe I could make the whole thing lighter by stripping down the camera a bit and using multiple smaller TECs distributed around the box.

I was trying to figure out how the high end CCD makers got efficient cooling sytems into such small packages. Looks like you have got there via your own R&D; lower power and smaller size of TEC for localised cooling.

[Gina]

It seems that "less is more" as far as TECs are concerned. But good thermal insulation is paramount with a whole box cooling setup. I think one TEC on the outside with a small finned heat sink with fan immediately inside would be best - well that's what I did anyway. You need to circulate cool air inside the box to cool the camera, though bolting the camera directly to the box with the tripod bush will help because the tripod bush is attached to the main metal frame inside the camera.

I guess it would help if you removed the back of the camera. That would help the cool air get inside. It will work without the display or buttons on the back. Save a bit of weight too. The front case moulding isn't necessary either if you can hold the battery connector and battery door in place to give power. The battery door operates a switch and you won't get any power unless this switch is held in (at least with the 1100D and I expect the 1000D is the same).

I'm not sure which other parts could be removed to save weight. There were several parts of the 450D which could be removed (as detailed in an extreme mod web site) but I found my 1100D wouldn't work without the flash unit intact, for instance.

When I was looking into whole camera cooling I wanted to remove the flash unit and front moulding to save weight and size and avoid needing a hole for the lumpy bit on the flash housing.

[Gina]

Peltier TEC manual control circuit.

[Reggie]

Ever thought of insulating with aerogel?

http://www.buyaerogel.com/

very expensive it would appear but they do different types, the 'spaceloft blanket' and the granules look like a cost effective way of doing things.

[Gina]

WOW!! That sure is expensive! Extremely expensive is how I would describe it

Insulation isn't a problem for me - my main problem is moisture and the resultant condensation and icing of the sensor. I need for find a way of sealing the camera and then absorbing any remaining moisture. I'm giving this a lot of thought. I might even change tack and go for a plastic box to put the camera in. That would make providing sealed connections much easier.

[nightvision]

My ideas for combating moisture include having a door in the cooler box so I can load up freshly charged desiccant. Could still have a challenge around the clip filter and sensor face so I may try desiccant between the filter and the coma-corrector.

[mindburner]

this is a great post. Lots of good info and advice contained within. It's amazing the cooling you have achieved Gina. ALthough you have problems, it seems well worth the effort. I was beginning to wonder if it was worth all this cooling DIY as I was getting ok images with the modded canon, but after being out in the hot evenings, the thermal noise was just too bad. So modding seems very worthwhile.

I have been really struggling with the alu box design cooler that I have built. I have another post about the peltier coolers that I have been using, but suffice to say they are not cooling the alu box at all. Like not even a degree! The box is made from 1mm alu plate, bent and formed into shape. Pop riveted and sealed up. Then covered in polystyrene foam (3cm)

I have come to the conclusion that I need a better material to make the box from. Or try a non peltier cooling system.

[russellhq]

What about an additional TEC to catch the moisture? As moisture tends to go for the coldest parts then that might help? Desiccant should help too.

[Gina]

My ideas for combating moisture include having a door in the cooler box so I can load up freshly charged desiccant. Could still have a challenge around the clip filter and sensor face so I may try desiccant between the filter and the coma-corrector.

That's a good idea And yes, clip filter and sensor face are a problem. May be possible to use silicone grease. But I have another idea for mine and that's to seal to the outside of the focal reducer, which I shall always be using. More of which later...

[Gina]

this is a great post. Lots of good info and advice contained within. It's amazing the cooling you have achieved Gina. ALthough you have problems, it seems well worth the effort. I was beginning to wonder if it was worth all this cooling DIY as I was getting ok images with the modded canon, but after being out in the hot evenings, the thermal noise was just too bad. So modding seems very worthwhile.

Thank you Yes, I found the same thing the other night without the cooling on. And yes, it's taking a long time but should be worth it in the end

I have been really struggling with the alu box design cooler that I have built. I have another post about the peltier coolers that I have been using, but suffice to say they are not cooling the alu box at all. Like not even a degree! The box is made from 1mm alu plate, bent and formed into shape. Pop riveted and sealed up. Then covered in polystyrene foam (3cm)

I have come to the conclusion that I need a better material to make the box from. Or try a non peltier cooling system.

Your ali box seems the same as others have used but copper would be better as it's twice as good at conducting heat. Have you got a finned heatsink and fan inside?

[Gina]

What about to catch the moisture? As moisture tends to go for the coldest parts then that might help? Desiccant should help too.

An additional TEC is a very interesting idea - it could be run colder so that it would catch any moisture before the sensor Yes, I'm using desiccant of course.

[Reggie]

WOW!! That sure is expensive! Extremely expensive is how I would describe it

Insulation isn't a problem for me - my main problem is moisture and the resultant condensation and icing of the sensor. I need for find a way of sealing the camera and then absorbing any remaining moisture. I'm giving this a lot of thought. I might even change tack and go for a plastic box to put the camera in. That would make providing sealed connections much easier.

Kind of, you can actually make your own www.aerogel.org has instructions.

I was really looking at the aerogel blanket for you guys, which is considerably cheaper than the blocks, the small pellets are cheaper than the blocks too!

[Gina]

I've come to the conclusion that putting the camera in a plastic box is the easiest way of providing a sealed casing. It also makes it much easier to provide connections to the camera, TECs, various circuitry and water pipes. As for sealing the light path, the easiest way would be to seal to the outer casing of the focal reducer. The box needs to clear the projection for the flash unit as my 1100D won't work with the flash unit removed - that is easier than cutting an odd shaped hole and trying to seal that. I expect to be always using the FR/FF with this camera and scope. Polyether foam can be used to form the seal.

The photo shows the camera as it is now with the FR/FF attached and an outline of a possible plastic case added.

[nightvision]

Looking cool :(. what about a small venting door and a small heater embedded in desiccant. You could charge the desiccant prior to a session, close the vent, turn off the heater and switch on the cooler.

I will be doing something like that.

[Gina]

Looking cool . what about a small venting door and a small heater embedded in desiccant. You could charge the desiccant prior to a session, close the vent, turn off the heater and switch on the cooler.

I will be doing something like that.

That's an idea I certainly think some way of either replacing or recharging the desiccant would be a good idea. The box I'm thinking of will have plenty of room inside.

[yesyes]

I've bought this silica gel:

http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&item=270968731230&ssPageName=ADME:X:AAQ:GB:1123

It is normally orange and changes to green when it's too wet.